Study On Opyimal Operation Strategies For Active Distribution Network

Future distribution network with high penetration of distributed generations (DGs) will present high uncertainty and high complexity and variability of operation conditions, due to the uncertainty and fluctuation of renewable energy generations. Active distribution network technology is proposed to provide a series of solutions to the challenges of large scale of renewable energy, which implement active control and coordinated optimization to distributed generations, energy storages and etc., via advanced communication technology and power electronic technology, to improve the distribution networks’ability of absorbing renewable energy generations. However, most existing distribution network optimization and control strategies are established based on deterministic optimal power flow (OPF) model and real-time control under multi-time scale framework, which weaken adverse impact of grid-connected DGs by’stepwise refining’and implement regulations passively after distribution network security constraints violations and can’t realize active control and real-time optimization. To overcome the drawbacks of existing studies, this thesis has done three levels’study of distribution network online power flow analysis technology, active and reactive power coordinated optimization technology for distribution network including DGs and robust reactive power optimization technology considering the uncertainty of DGs’output power under the framework of active distribution network. The main contributions of this thesis is as follows:1. The technological background and connotation of active distribution network is introduced, the latest achievements in the field of uncertain power flow analysis and optimal operation of distribution network is reviewed and summarized, and develop trends of active distribution network technology is discussed.2. The demand and necessity of online deterministic and uncertain power flow analysis for distributed network in the visual angle of active distributed network is discussed and elaborated. The calculation method of three phase admittance matrix is derived, three phased power flow equations is acquired for unbalanced distribution network, and based on this, an approximated linear power flow algorithm is developed. The basic theory and process of analyzing uncertain problem via affine arithmetic is introduced and a kind of multi-stage affine arithmetic based interval power algorithm is proposed, witch improves the accuracy and speed of interval power flow analysis a lot by eliminating the inaccurate affine division and reducing the numbers of affine multiplication. In addition, the newly developed interval power flow method could also provide a high precision quantitative mapping from the power injections of distributed generation to the complex bus voltage.3. Active and reactive power coordinated optimization strategy is developed for distribution network operation. The branch flow based and bus injection based coordinated optimization model are established respectively for balanced and unbalanced distribution network. A second order cone is designed for the branch flow based model, which solves the optimization efficiently and globally by relaxing the primal non-convex model to a convex second order cone programming model. A trust region sequential linear programming method is proposed to solve the bus injection based model, which improve the convergence accuracy and speed of primal sequential linear programing method by the trust region technology.4. Robust reactive power optimization strategy linear adaptive robust reactive power optimization strategy is developed considering the uncertainty of distributed generations’output power, which guarantees no constraints violations occur over the whole uncertainty set of distributed generations’active output power. A robust reactive power optimization model considering uncertainty is established, a benders decomposition type cutting plain method is proposed to solve the robust minimax problem. Based on this, a linear adaptive robust reactive power optimization model is developed and a kind of two stage approximated solver is proposed, which regulates the reactive power output of the distributed generations adaptively according to the monitored active power output, tracing the optimal operation point of the distribution network in real time, thus reducing the conservatism of the non-adaptive robust optimization method greatly.